Page 60 - Spring2022

HEADPHONE SOUND QUALITY
The DF target was not seriously challenged until Lorho (2009) reported 80 listeners (25% audio engineers, 25% music students, and 50% naive listeners) on average pre- ferred a significantly modified version of the DF target where its main feature, a wide 12 dB peak at 3 kHz, was reduced to just 3 dB. This paper sparked new interest to find better alternative headphone target curves to the ones recommended in the current headphone standards.
Semireflective Field Headphone Equalizations (2012 to Present)
Because stereo recordings are optimized for reproduc- tion through loudspeakers in semireflective rooms, they should sound best through headphones that emulate this sound field. Sank (1980) made similar proposals three decades earlier but never conducted formal listening tests that compared these targets with the DF target.
Loudspeakers with flat on-axis and smooth off-axis fre- quency responses tend to produce the highest subjective ratings in formal listening tests (Toole, 2018). When placed in a typical room, they produce a uniform quality of direct, early, and late reflected sounds that in summa- tion produce the steady-state in-room response of the loudspeaker. Due to the frequency-dependent directiv- ity of the loudspeaker and absorption characteristics of the room, the in-room response will not be flat like the FF response nor the same as the DF response where the room absorption has been removed. Instead, the in-room response gently falls about 1 dB per octave from 20 Hz to 20 kHz.
Fleishmann et al. (2012) reported the first formal listen- ing test results where three SRF headphone targets were evaluated. The targets were based on measurements of the steady-state in-room response of a 5.1-channel loud- speaker setup in a standard listening room and then equalized by three expert listeners to match the timbre of the speakers. Two of the SRF targets were found to be slightly preferred to the DF target, depending on the music programs. Other targets included the Lorho target, a flat target, and three unequalized headphones that generally received lower ratings than the two SRF targets. Unfortunately, no measurements or details of the loudspeakers and the three SRF targets were given. The conclusions were that the SRF targets were equal to or better than the DF target, but the Lorho target was not.
A similar study (Olive et al., 2013a) reported evidence that listeners strongly preferred headphones equalized to SRF targets to, in descending order of preference, two DF tar- gets (Möller et al., 1995); two high-quality headphones; the Lorho target; and the FF target. The trained listeners described both DF targets as having too much emphasis in the upper midrange (2-4 kHz) and lacking bass. The Lorho target had too little energy at 2-4 kHz, which made instruments sound “muffled and dull.” The FF target was strongly criticized for its strong emphasis between 2 and 4 kHz, lack of bass, and harsh and nasal colorations. Listen- ers described the highest rated the SRF target as having
“good bass with an even spectral balance.” The measured frequency responses of the headphone targets correlate to and confirm listeners’ descriptions of their sound quality (see Olive et al., 2013b, Figure 2). The highest rated target curve in this study soon became known in the audio indus- try as the Harman target curve and is widely influencing the design, testing, and review of headphones.
Do Listeners Agree on What Makes a Headphone Sound Good?
Although the initial test results of the Harman target curve were encouraging, they were based on a small sample of 10 trained listeners. To better understand if certain demographic factors influence the acceptance of the curve, it was tested using a larger number of listeners from a broad range of ages, listening experiences, and geographic regions.
The target curve was benchmarked against three head- phones considered industry references at the time in terms of sound quality or commercial sales (Olive et al., 2014). They ranged in price from $269 to $1,500 and included dynamic and magnetic planar transducer designs. A total of 283 listeners participated from four different countries (Canada, United States, Germany, and China) and included a broad range of ages, listen- ing experiences, and genders. Most of the participants were Harman employees.
A novel virtual headphone test methodology allowed controlled, rapid, double-blind comparisons among the different headphones. Virtual versions of the different headphones were reproduced over a single high-quality replicator headphone by equalizing it to match the mea- sured frequency response of each headphone. This removed
60 Acoustics Today • Spring 2022